Secure Embedded Data Schemes for User Adaptive Multimedia Presentation

Authors

  • Neelu Sinha Fairleigh Dickinson University, Dept. of Computer Science, Mathematics and Physics

Abstract

In this Digital/Internet Age, digital multimedia holds an unlimited potential, and virtually all forms of media content, including books, video games, music and software are now available for digital distribution. Digital multimedia libraries, comprising a large amount of such digital media (the so-called Digital Intellectual Property), in the form of images, video, audio and graphics are rapidly growing. Also, due to the unprecedented growth of the World Wide Web, vast amounts of multimedia data is readily available leading to an explosion of multimedia and hypermedia database creation and sharing. Digital information embedding techniques for various types of media, for a variety of applications including digital libraries, museum cataloging, medical and healthcare industries, digital preservation systems, educational systems and personalization systems, are of significant interest in two areas. Firstly, these are useful for the realization of efficient database indexing schemes and customization, which in turn lead to efficient tools for the organization, retrieval, adaptive presentation, and distribution of digital media content. Secondly, these are useful in developing tools to protect, detect and verify ownership and/or usage rights for the Digital Intellectual Property and also the tracking of these in the distribution medium. In either of these applications, information embedding schemes which allow for a detailed level of source description, and which are robust to some of the distortions encountered in the distribution medium (e.g., JPEG compression for images), are particularly attractive. This paper presents an overview of multimedia presentation adaptation through the use of robust information descriptors and a novel information embedding technique for digital images that allows for significantly higher information throughput and increased robustness compared to many of the existing techniques.

Downloads